Alzheimer's disease and related dementias (ADRD) has a major impact on how many elderly are cared for, the cost of such care to public programs (Medicare, Medicaid) and to families. Poor cognitive functioning is an important risk factor in functional decline, institutionalization and mortality. The number of persons with ADRD is rising as mortality from other causes is delayed and the population ages. It remains unclear whether demographically adjusted prevalence is changing. The cost of ADRD is staggering from small convenience samples and cross sectional designs. There is much to be learned from longitudinal analysis of a nationally representative data bases that has become available recently, the National Long Term Care Survey (NLTCS). The specific aims of the proposed work are: (1) Measure the age and gender adjusted prevalence of ADRD in 1984, 1989, 1994 and 1999 by reconciling Medicare claims based diagnosis of ADRD with patients identified using the survey instrument of NLTCS; (2) Ascertain the important determinants of prevalence over time, including decreasing mortality from cardiovascular diseases and receipt of life saving interventions; (3) Determine whether persons with ADRD receive less intensive treatment for major conditions other than ADRD (e.g., selected cancers, coronary heart disease) relative to otherwise similar disabled elderly persons; (4) Measure the cost of ADRD to Medicare and Medicaid other government agencies, private payers and out-of-pocket the cost of ADRD to Medicare and Medicaid other government agencies, private payers and out-of-pocket expenditures for personal health services for ADRD, in 1984, 1994 and 1999m and project cost trends for these public programs based on these estimates. To compute such cost estimates, we will account for two factors often overlooked in such projections: (1) the impact of ADRD on the treatment of other health conditions and (2) techniques that such public programs as Medicare would incur other costs from other diseases if ADRD were eliminated. The NLTCS will allow this project to provide the best information to date on the adjusted prevalence of ADRD, what factors are influencing same, whether persons with ADRD are treated differently given acute and chronic health conditions, and how ADRD will impact future medical costs.